Doctoral Student in Machine Learning for Robotic Manipulation

This PhD position at KTH Royal Institute of Technology is focused on machine learning for robotic manipulation within Associate Professor Florian Pokorny's research group. The project aims to develop innovative machine learning approaches that integrate large-scale, real-world robotic manipulation experiments and benchmarking into a robust, statistically reliable data-driven pipeline. The ultimate goal is to enable robots to learn new manipulation skills efficiently and reliably. A key component of the research involves leveraging CloudGripper, the lab’s large-scale, remotely accessible cloud robotics platform, which features a variety of robotic arm workcells. The project emphasizes the creation of internationally relevant datasets, benchmarks, and research results to establish a reliable ground truth for robotic manipulation research. The successful candidate will join a dynamic and international research environment, collaborating with industry partners and leading universities worldwide. The position offers a monthly salary according to KTH’s doctoral student salary agreement, along with employee benefits. Applicants must have a second cycle degree (such as a master's) or equivalent, with at least 240 higher education credits (including 60 at the second-cycle level), and demonstrate English proficiency equivalent to English B/6. Selection will be based on academic qualifications, motivation, ability to work independently and collaboratively, and personal skills. The position is full-time, with employment for up to four years, and may require a security clearance. Applications must be submitted through KTH’s recruitment system by the specified deadline and should include a CV, application letter, and relevant publications. KTH is committed to equality, diversity, and providing a creative, supportive environment for doctoral students.